Follow-up through Middle School of a Randomized Study of Public Prekindergarten Project Summary (30 lines) The experimental evidence for positive effects of prekindergarten programs comes from studies of preschool programs that are more intensive than the current public prekindergarten models and which were not implemented on the same scale as typical contemporary public pre-k. Though it is not widely recognized, there are in fact no well-controlled experimental studies of typical public prekindergarten programs with longitudinal follow-up that would reveal whether such programs produce the expected long-term effects. This application proposes a follow-up through middle school of a large randomized control trial of a state-funded public prekindergarten program, specifically the Tennessee Voluntary Pre-K Program (TN-VPK). TN-VPK serves children eligible for the federal free or reduced price lunch programs, requires a licensed teacher and an aide in each classroom, a maximum class size of 20 children, and a state-approved curriculum. This application proposes a follow-up from 4th grade through 8th grade of 3008 children who were randomly assigned to participate or not in TN-VPK. Data were collected annually from the state database for all children and individually for a subsample of 1076 children with parental consent. The findings of the TN-VPK study to date have shown what other longitudinal studies of more intensive early childhood programs have generally found: a wide range of positive outcomes immediately after program participation and mixed effects over a longer term. The effects on achievement measures often fade out, but behavioral/non-cognitive effects such as lower grade retention, higher graduation rates, decreased delinquency, and the like begin to emerge. The procedural aims of the research proposed in this application are to (a) follow the full randomized sample (N=3008) through middle school with annual data collection on achievement and behavioral/non-cognitive outcomes from the Tennessee state education data system, and (b) collect additional more extensive data annually from the children and their parents and teachers for a subsample of approximately 700 children. During middle school the instructional material becomes more challenging, children are required to take greater responsibility for their learning, and problem behaviors often emerge, including many of the behavioral/non-cognitive effects others have found to be influenced by early childhood. The major analyses required to address the specific aims of this research fall into three categories: (a) estimation of the main effects o TN-VPK participation on late elementary and middle school outcomes; (b) exploration of the extent to which TN-VPK effects on some of those outcomes (self-regulation, school engagement, peer relations) that are evident by the end of the elementary grades mediate the effects on achievement, other academic outcomes, and problem behavior outcomes at the end of middle school; and (c) investigation of the extent to which TN-VPK main effects and/or mediated effects on the various outcomes are moderated by the quality of the schools attended or demographic characteristics of the students.